INDEX

Explanations

proper nouns, particularly names related to organizations or locations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

roman

-0.20

utter

-0.17

_epi

-0.16

avin

-0.15

istros

-0.15

metatable

-0.15

undos

-0.15

 Labs

-0.15

_hdl

-0.15

stanov

-0.14

POSITIVE LOGITS

ammu

0.21

aved

0.17

hoot

0.17

at

0.16

heel

0.16

hill

0.15

hal

0.15

hangi

0.15

omor

0.15

han

0.14

Activations Density 0.041%

proper nouns, particularly names related to organizations or locations

No Comments

No Known Activations